This project is concerned with the development of a new drug development paradigm that focuses on the use of preclinical studies in animals to derive predictive pharmacokinetic [PK] - pharmacodynamic [PD] models that can aid in the rational selection of drug candidates, and be used to predict human PK-PD characteristics. A unique feature of the proposal is that the drug's PK-PD properties in target tissues, in this case brain and brain tumors, are integral to the drug development plan. The proposed drug development strategy will be demonstrated for a series of epidermal growth factor receptor [EGFR] inhibitors, a class of small molecular weight compounds that are represenative of targeted anticancer drugs, an emerging anticancer therapeutic strategy. The motivation for this new drug development approach is based on limitations of current semi-empirical anticancer drug development approaches, which rely on a plethora of drug efficacy studies that lack predictive capability on how drugs should be used in humans. Moreover, EGFR inhibitors are representative of the problems of defining optimal doses for targeted therapies in cancer patients, since these classes of drugs often do not possess readily identified dose-limiting toxicites that typically guide the use of standard cytotoxics. The PK-PD driven approach will rely on the use of hybrid PK-PD models that enables a target tissue [i.e. brain tumors] to be represented in physiological terms, which accomodates a mechanistic depiction of drug disposition, and facilitates model scale-up to humans. The drug development scheme will be implemented in brain tumor models that differ only in EGFR status, one being wild-type and the other the vIII mutant, which is associated with drug sensitivity, which will allow us to test the broader applicability of the approach to other brain tumor models whose molecular characteristics are known. A synopsis of the Specific Aims are 1) screen new EGFR inhibitors for brain accumulation in a cassette dosing format;2) determine in vitro PD endpoints associated with EGFR signaling pathways;3) develop hybrid PK-PD models in mice;and 4) evaluate drug efficacy in a range of brain tumor types based on the use of PK-PD equivalent dosing regimens. The proposed drug development package relies on a drug's in vivo pharmacological behavior to screen, select, and subsequently design therapeutic regimens based on model-predicted PK-PD endpoints. It is believed that this approach offers the best opportunity to develop drugs in a rational and optimal manner.